From: Chen Gang <gang.chen@asianux.com>
To: Peter Ujfalusi <peter.ujfalusi@ti.com>, sameo@linux.intel.com
Cc: Tony Lindgren <tony@atomide.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: [Suggestion] drivers/mfd/twl*: about power_count in twl6040_remove
Date: Sat, 16 Feb 2013 17:07:55 +0800 [thread overview]
Message-ID: <511F4C6B.4030805@asianux.com> (raw)
if necessary to treat 'power_count' as a counter
when power_count is more than 1, we do not really power off it.
this may cause issue:
after finish calling twl6040_remove, another modules still use it.
suggest to give additional check, or synchronisation for other modules.
else (not necessary to treat it as a counter)
better to use it as a boolean value.
it seems 'power_count' may still have chance to be more than 1:
if it is true, we still may can not power off it in twl6040_remove.
another question:
may 'power_count' be a large number (which may cause value overflow) ?
thanks.
gchen.
------------------------------------------------------------------------------
relative calls:
drivers/mfd/twl6040.c:212: twl6040_power(twl6040, 0);
drivers/mfd/twl6040.c:215: twl6040_power(twl6040, 1);
drivers/mfd/twl6040.c:707: twl6040_power(twl6040, 0);
drivers/input/misc/twl6040-vibra.c:100: twl6040_power(info->twl6040, 1);
drivers/input/misc/twl6040-vibra.c:128: twl6040_power(info->twl6040, 0);
sound/soc/codecs/twl6040.c:911: ret = twl6040_power(twl6040, 1);
sound/soc/codecs/twl6040.c:926: twl6040_power(twl6040, 0);
drivers/clk/clk-twl6040.c:51: ret = twl6040_power(twl6040_clk->twl6040, 1);
drivers/clk/clk-twl6040.c:64: ret = twl6040_power(twl6040_clk->twl6040, 0);
drivers/mfd/twl6040.c:
244 int twl6040_power(struct twl6040 *twl6040, int on)
245 {
246 int ret = 0;
247
248 mutex_lock(&twl6040->mutex);
249
250 if (on) {
251 /* already powered-up */
252 if (twl6040->power_count++)
253 goto out;
254
255 if (gpio_is_valid(twl6040->audpwron)) {
256 /* use automatic power-up sequence */
257 ret = twl6040_power_up_automatic(twl6040);
258 if (ret) {
259 twl6040->power_count = 0;
260 goto out;
261 }
262 } else {
263 /* use manual power-up sequence */
264 ret = twl6040_power_up_manual(twl6040);
265 if (ret) {
266 twl6040->power_count = 0;
267 goto out;
268 }
269 }
270 /* Default PLL configuration after power up */
271 twl6040->pll = TWL6040_SYSCLK_SEL_LPPLL;
272 twl6040->sysclk = 19200000;
273 twl6040->mclk = 32768;
274 } else {
275 /* already powered-down */
276 if (!twl6040->power_count) {
277 dev_err(twl6040->dev,
278 "device is already powered-off\n");
279 ret = -EPERM;
280 goto out;
281 }
282
283 if (--twl6040->power_count)
284 goto out;
285
286 if (gpio_is_valid(twl6040->audpwron)) {
287 /* use AUDPWRON line */
288 gpio_set_value(twl6040->audpwron, 0);
289
290 /* power-down sequence latency */
291 usleep_range(500, 700);
292 } else {
293 /* use manual power-down sequence */
294 twl6040_power_down_manual(twl6040);
295 }
296 twl6040->sysclk = 0;
297 twl6040->mclk = 0;
298 }
299
300 out:
301 mutex_unlock(&twl6040->mutex);
302 return ret;
303 }
304 EXPORT_SYMBOL(twl6040_power);
...
702 static int twl6040_remove(struct i2c_client *client)
703 {
704 struct twl6040 *twl6040 = i2c_get_clientdata(client);
705
706 if (twl6040->power_count)
707 twl6040_power(twl6040, 0);
708
709 if (gpio_is_valid(twl6040->audpwron))
710 gpio_free(twl6040->audpwron);
711
712 free_irq(twl6040->irq_ready, twl6040);
713 free_irq(twl6040->irq_th, twl6040);
714 regmap_del_irq_chip(twl6040->irq, twl6040->irq_data);
715
716 mfd_remove_devices(&client->dev);
717 i2c_set_clientdata(client, NULL);
718
719 regulator_bulk_disable(TWL6040_NUM_SUPPLIES, twl6040->supplies);
720 regulator_bulk_free(TWL6040_NUM_SUPPLIES, twl6040->supplies);
721
722 return 0;
723 }
-----------------------------------------------------------------------------------
next reply other threads:[~2013-02-16 9:08 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-02-16 9:07 Chen Gang [this message]
2013-02-18 9:16 ` [Suggestion] drivers/mfd/twl*: about power_count in twl6040_remove Peter Ujfalusi
2013-02-18 9:21 ` Chen Gang
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=511F4C6B.4030805@asianux.com \
--to=gang.chen@asianux.com \
--cc=linux-kernel@vger.kernel.org \
--cc=peter.ujfalusi@ti.com \
--cc=sameo@linux.intel.com \
--cc=tony@atomide.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.